The present invention relates to a method and apparatus for adaptively coding video signals; and, more particularly, to a method and apparatus capable of adaptively the coding video signals by way of predicting intra DC and AC coefficients, separately.
In digitally televised systems such as video-telephone, teleconference and high definition television systems, a large amount of digital data is needed to define each video frame signal since the video frame signal comprises a sequence of digital data referred to as pixel values. Since, however, an available frequency bandwidth of a conventional transmission channel is limited, in order to transmit the large amount of digital data therethrough, it is necessary to compress or reduce the volume of the data through the use of various data compression techniques, especially in the case of such low bit-rate video signal encoders as video-telephone and teleconference systems.
Among various video compression techniques, the so-called hybrid coding technique, which combines temporal and spatial compression techniques together with a statistical coding technique, is known to be most effective. Most hybrid coding techniques employ a motion compensated DPCM (Differential Pulse Coded Modulation), two-dimensional transform, e.g., DCT (Discrete Cosine Transform), quantization of transform coefficients, and an entropy coding, e.g., VLC (Variable Length Coding).
In the conventional MPEG-4 (Moving Picture Expert Group-4) coding scheme, an intra-mode input video signal is divided into blocks of Mxc3x97N, e.g., 8xc3x978, pixels and each block is transformed to a set of quantized coefficients, i.e., a DC and a plurality of AC quantized coefficients, through the use of two-dimensional transform, e.g., DCT and quantization thereof, M and N being positive integers. A set of quantized coefficients of a block is then DPCM coded based on a set of quantized coefficients of one of previously coded blocks in the frame.
Referring to FIG. 1, there is shown a portion of blocks of an intra-frame. In MPEG-4, a reference block RB of a current block is determined among a left and an upper blocks of the current block based on DC coefficients of the left and the upper and a left-upper blocks of the current block. If a current block is 108, a reference block RB thereof is determined as:
If |D3xe2x88x92D1| less than | less than D1xe2x88x92D2|, then
RB is the upper block 104,
else
RB is the left block 106xe2x80x83xe2x80x83(Eq. 1)
wherein D1, D2 and D3 are DC quantized coefficients of the blocks 102, 104 and 106, respectively.
In other words, if the horizontal correlation between the DC quantized coefficients of the previously coded blocks is greater than the vertical correlation, i.e., |D1xe2x88x92D2| less than |D3xe2x88x92D1|, the previously coded left block 106 is selected as the RB for the current block 108; and if otherwise, the previously coded upper block 104 is selected. Thereafter, a DPCM DC coefficient for the current block 108 is generated by calculating the difference between the DC quantized coefficient of the RB and that of the current block 108.
According to the conventional MPEG-4 coding method, the RB determined based on the DC quantized coefficients of the previously coded blocks is also used in predicting the AC quantized coefficients of the current block without considering the correlations between the AC quantized coefficients of the previously coded blocks, which may result in degraded coding efficiency.
It is, therefore, an object of the present invention to provide a method and apparatus capable of adaptively coding intra-video signals by separately predicting intra DC and AC coefficients.
In accordance with the present invention, there is provided a method for encoding a current block of a video signal based on P previously coded blocks, each of the current block and the previously coded blocks having Mxc3x97N coefficients of a DC and a plurality of AC coefficients and P being an integer greater than 2, comprising the steps of:
(a) finding a DC reference block among the previously coded blocks based on a correlation of the DC coefficients of the previously coded blocks;
(b) setting a DC coefficient of the DC reference block as a DC reference coefficient;
(c) determining one of the previously coded blocks as an AC reference block based on a correlation of selected AC coefficients of the previously coded blocks;
(d) generating AC reference coefficients based on AC coefficients of the AC reference block; and
(e) encoding coefficients of the current block based on the DC and the AC reference coefficients.
In accordance with the present invention, there is provided an apparatus for encoding a video signal, the video signal is divided into a plurality of blocks of Mxc3x97N pixels, comprising: a transform block for transforming the video signal on a block-by-block basis to thereby a set of Mxc3x97N transform coefficients for each block, wherein the transform coefficients includes a DC and a multiplicity of AC transform coefficients; a quantization block for quantizing the set of transform coefficients for each block to provide a set of Mxc3x97N quantized DC and AC coefficients; a memory for storing the set of quantized coefficients for each block; a prediction block for determining a DC reference block and an AC reference block of a current block based on quantized DC coefficients and quantized AC coefficients of previously coded blocks to thereby provide a set of DPCM (differential Pulse Coded Modulation) DC and AC coefficients of the current block based on quantized DC and AC coefficients of the DC and the AC reference blocks; and an entropy coding block for entropy coding the DPCM DC and the DPCM AC coefficients.